The 17 alpha-alkylated properties of methanedienone do make it liver toxic, but this, I believe, is overstated as most of the evidence of its toxicity comes from studies on individuals and not from studies on large groups of dianabol-using bodybuilders. One study on rats (1) showed that regardless of dose or time of administration, dianabol produces changes in enzymatic activity, which leads to hypertrophy of hepatocytes; which basically shows that dianabol is toxic to the liver. But in another study (2) Nerobol (Russian Dianabol) was found to favour a rapid normalisation of functional and metabolic disorders of the liver, which contradicts the earlier evidence. This shows that the whole idea of dianabol being dangerous is in no way as bad as some would make out.